SuperLU Distributed 9.0.0
gpu3d
batch_block_copy.h
Go to the documentation of this file.
1#ifndef __BATCH_BLOCK_COPY_H__
2#define __BATCH_BLOCK_COPY_H__
3
4#include <cublas_v2.h>
5
6#ifdef __cplusplus
7extern "C" {
8#endif
10 cudaStream_t stream, int* m_batch, int* n_batch, int max_m, int max_n,
11 float** dest_ptrs, int* dest_ld_batch, float** src_ptrs, int* src_ld_batch,
12 int ops
13);
14
16 cudaStream_t stream, int* m_batch, int* n_batch, int max_m, int max_n,
17 double** dest_ptrs, int* dest_ld_batch, double** src_ptrs, int* src_ld_batch,
18 int ops
19);
20
21#ifdef __cplusplus
22}
23#endif
24
25#ifdef __cplusplus
26inline int copyBlock_vbatch(
27 cudaStream_t stream, int* m_batch, int* n_batch, int max_m, int max_n,
28 float** dest_ptrs, int* dest_ld_batch, float** src_ptrs, int* src_ld_batch,
29 int ops
30)
31{
32 return scopyBlock_vbatch(
33 stream, m_batch, n_batch, max_m, max_n, dest_ptrs, dest_ld_batch,
34 src_ptrs, src_ld_batch, ops
35 );
36}
37
38inline int copyBlock_vbatch(
39 cudaStream_t stream, int* m_batch, int* n_batch, int max_m, int max_n,
40 double** dest_ptrs, int* dest_ld_batch, double** src_ptrs, int* src_ld_batch,
41 int ops
42)
43{
44 return dcopyBlock_vbatch(
45 stream, m_batch, n_batch, max_m, max_n, dest_ptrs, dest_ld_batch,
46 src_ptrs, src_ld_batch, ops
47 );
48}
49#endif
50
51#endif //__BATCH_BLOCK_COPY_H__
int scopyBlock_vbatch(cudaStream_t stream, int *m_batch, int *n_batch, int max_m, int max_n, float **dest_ptrs, int *dest_ld_batch, float **src_ptrs, int *src_ld_batch, int ops)
int dcopyBlock_vbatch(cudaStream_t stream, int *m_batch, int *n_batch, int max_m, int max_n, double **dest_ptrs, int *dest_ld_batch, double **src_ptrs, int *src_ld_batch, int ops)